Article
Automation & Control Systems
Atefeh Daemi, Bhushan Gopaluni, Biao Huang
Summary: In this article, we propose a novel transfer learning approach, called domain adversarial probabilistic principal component analysis (DAPPCA), to monitor processes with data from multiple distributions. DAPPCA automatically learns feature representations that are relevant across different operational modes and improves fault detection accuracy by transferring knowledge from previously known modes.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Automation & Control Systems
Zhiwen Chen, Chang Liu, Steven X. Ding, Tao Peng, Chunhua Yang, Weihua Gui, Yuri A. W. Shardt
Summary: A new method for monitoring and fault detection of multimode processes is proposed in the article, integrating K-means into just-in-time learning to build local models and addressing limitations of traditional canonical correlation analysis methods in handling processes with multiple operating points. Its effectiveness is demonstrated in an industrial benchmark process, showing better fault detection rate compared to conventional methods.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Automation & Control Systems
Jingxin Zhang, Donghua Zhou, Maoyin Chen
Summary: This article proposes a novel sparse principal component analysis algorithm with self-learning ability for multimode process monitoring. The proposed method learns the successive modes in a sequential fashion and remembers the learned knowledge by selectively slowing down the changes of important parameters for the previous modes. The algorithm has excellent interpretability and alleviates the catastrophic forgetting problem.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Chemical
Hairong Fang, Wenhua Tao, Shan Lu, Zhijiang Lou, Yonghui Wang, Yuanfei Xue
Summary: This paper proposes a new two-step dynamic local kernel principal component analysis method, which can handle the nonlinearity and the dynamic features simultaneously.
Article
Microbiology
Jinlin Zhu, Heqiang Xie, Zixin Yang, Jing Chen, Jialin Yin, Peijun Tian, Hongchao Wang, Jianxin Zhao, Hao Zhang, Wenwei Lu, Wei Chen
Summary: The study developed a statistical monitoring framework to predict and analyze individual health status based on the gut microbiome. It found that the gut microbiome is closely associated with health and identified the contribution of each bacterium to health. The framework can help to understand healthy microbiomes, unhealthy variations, and discover personalized therapy targets.
Article
Computer Science, Artificial Intelligence
Qingchao Jiang, Shifu Yan, Hui Cheng, Xuefeng Yan
Summary: This article proposes a local-global modeling and distributed computing framework for efficient fault detection and isolation in nonlinear plant-wide processes. With the use of stacked autoencoders and mutual information, dominant representations are extracted, neighborhood variables are determined, and monitors are established to achieve global monitoring systems. The feasibility of the method is demonstrated through application to the Tennessee Eastman (TE) and laboratory-scale glycerol distillation processes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Engineering, Chemical
Zhijiang Lou, Youqing Wang, Shan Lu, Pei Sun
Summary: This study proposes a novel robust PCA scheme called MRPCA, which adopts a difference selection mechanism for outlier samples in the offline training stage and an outlier detection mechanism for distinguishing outliers from fault data in the online monitoring stage. With these mechanisms, MRPCA achieves high fault detection rates and low false alarm rates in tests.
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH
(2021)
Article
Computer Science, Interdisciplinary Applications
Fan Guo, Bing Wei, Biao Huang
Summary: This paper proposes a soft sensor model based on GMVAE, utilizing GMM to extract multimodal characteristics and measuring similarity between new samples and historical data through MSKL divergence to establish a local model.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Engineering, Chemical
Yuan Li, Dongsheng Yang
Summary: LCPCA is a novel approach for monitoring the status of multimode processes without the need for prior knowledge, by dividing data into local components and applying posterior probability. It outperforms conventional PCA and LNS-PCA in fault detection rate based on numerical examples and the TE process.
CHINESE JOURNAL OF CHEMICAL ENGINEERING
(2021)
Article
Engineering, Chemical
Lin Xuan You, Junghui Chen
Summary: The study proposes a multi-local principal component analysis (ML-PCA) modeling strategy to monitor a non-linear process over a large operating region. ML-PCA automatically enhances the data of the clustered model and weakens those data of the other local models, demonstrating improved accuracy and effectiveness in process control.
CHEMICAL ENGINEERING SCIENCE
(2021)
Article
Automation & Control Systems
Xiangyin Kong, Zhiqiang Ge
Summary: This article proposes a novel monitoring framework for latent variable models using hierarchical feature extraction, Bayesian inference, and weighting strategy. The framework includes a deep PCA-ICA model for hierarchical feature extraction, Bayesian inference for transforming the features to posterior probabilities, and a weighting strategy for combining the probabilities into new probabilistic statistics. The proposed model is validated using the Tennessee Eastman process and the effectiveness of the deep hierarchical feature extraction structure is further analyzed.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Computer Science, Information Systems
Tian Cheng, Kunsong Zhao, Song Sun, Muhammad Mateen, Junhao Wen
Summary: With the rise of mobile devices, Android mobile apps have become indispensable in people's daily lives. This article proposes a novel method called KAL for cross-project defect prediction in Android mobile apps. By transforming and extracting features from commit instances, KAL achieves better performance compared to other comparative methods.
FRONTIERS OF COMPUTER SCIENCE
(2022)
Article
Automation & Control Systems
Wei Fan, Qinqin Zhu, Shaojun Ren, Liang Zhang, Fengqi Si
Summary: This article proposes a multistep dynamic predictive monitoring scheme that can handle measurement noise, and introduces a dynamic index to detect dynamic anomalies.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2022)
Article
Environmental Sciences
Yiran Liu, Jian Wang, Cheng Yang, Yu Zheng, Haipeng Fu
Summary: This study proposed a high-accuracy TEC prediction model based on machine learning, utilizing methods such as principal component analysis, solar activity parameters, and spatial interpolation. The model showed high consistency with observed values and outperformed the traditional IRI model.
Article
Automation & Control Systems
Kai Wang, Zhenli Song
Summary: This article proposes a novel privacy-preserving cross-plant process monitoring framework using federated learning. It reduces the data dimension with a new distributed principal component analysis (PCA) method and exchanges model parameters between different plants using a secure federated learning protocol. The superiority of this framework is validated through numerical simulations and real industrial case studies.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)